A range of criteria have been created to evaluate the appropriateness of gastrointestinal (GI) endoscopy. The most widely recognized were set by the American Society for Gastrointestinal Endoscopy and the European Panel of Appropriateness of Gastrointestinal Endoscopy (EPAGE), updated EPAGE-II criteria. Nevertheless, the application of general rules is not sufficient, especially in hospitalized patients. The complexity, singularity, and dynamic clinical evolution demands the proactive involvement of an experienced endoscopist to make the decision whether or not to perform the exam.

Our aim was to analyze the appropriateness of endoscopic exam requests in hospitalized patients over the course of one year.

A prospective observational study was conducted in an endoscopic unit of a tertiary hospital center. The requested exams were triaged by a gastroenterologist after reviewing the patient’s medical record, assessing clinical status and complementary diagnostic tests. In cases of unclear indication, and no absolute contraindication, the decision to proceed was discussed with the referring physician. 

A total of 184 requests for inpatient endoscopy and/or colonoscopy were received during 1 year (table 1). 154 (84%) of the requests were thought to be appropriate. 12 (6%) were cancelled during triage: 2 were patients on best supportive care, 1 died, 3 had other diagnosis which plausibly justified the clinical situation, 2 had recent endoscopic exams, and 4 presented no clinical/analytical evidence of GI bleeding). 18 (10%) requests were initially postponed (8 were awaiting imaging results; 5 had poor clinical status and 5 respiratory distress) and ended up being cancelled (8 had another diagnosis in imaging tests, 8 maintained clinical deterioration contraindicating exam and 3 died). In total 30 (16%) of the requests were cancelled. 

The most frequent exam requested was endoscopy, and the most common indication was suspected GI neoplasia. From a total of 58 suspected GI neoplasms, 11 were confirmed as gastric neoplasia, which was the most common finding.

None of the patients with a cancelled exam was diagnosed with a GI disease during follow up.

Mortality rate was considerably higher in the group with cancelled exams (33% vs 11% during hospitalization, 50% vs 32% during follow up, 15% vs 1.8% within 30 days after discharge). Mean age was similar between groups. This findings reveal a frail group of patients, with little likelihood of benefiting from an invasive exam. 

16% of the requests were considered inappropriate, which allowed to avoid 30 unnecessary exams.

Developing standardized indications for endoscopic exams in inpatient care is difficult, and an individualized approach is critical in order to reach an equilibrium between overusage and overlooking, while considering the principle of primum non nocere.